As one of the methods for testing semiconductor devices such as a very large scale integrated circuit (VLSI), a charged particle beam test system may be used to monitor voltages of internal nodes of the VLSI. A charged particle beam, such as an electron beam, is irradiated onto the surface of an integrated circuit under test (DUT) and the resultant secondary electron which represents the voltages and other states of the DUT is detected, thereby obtaining a voltage contrast image of the DUT on a display. A charged particle beam test system is advantageous for testing high density and complex semiconductor chips because it can test such semiconductor chips without physically contacting the chips for probing input/output signals.
In a modern charged particle beam test system, a charged particle beam tester and an IC tester interact with each other to analyze the performance of a DUT. As is well known in the art, an IC tester provides a test pattern signal (also called a test vector) to a DUT and compares the resultant output signal from the DUT with an expected value signal. Such a comparison is made at a timing predetermined by a comparison enable signal provided in a test program. If the output signal from the DUT is different from the expected value signal, the IC tester generates a fail signal.
In a charged particle beam test system having an IC tester, a DUT is placed on an X-Y stage of the charged particle beam column. The DUT is provided with a test pattern signal from the IC tester and the resultant output signal of the DUT is compared with the expected value signal as noted above by the IC tester. At the same time, the specified portion of the DUT which is being tested by the IC tester is also tested by the charged particle beam tester by being irradiated with the charged particle beam and the resultant secondary electron is detected. In this manner, in a charged particle beam test system, test result data is acquired concerning the DUT's output signal produced in response to the test pattern signal as well as concerning the potential contrast of the inner nodes of the DUT relative to the test pattern, resulting in a highly sophisticated and efficient evaluation of the DUT performance.
An example of such a charged particle beam test system in the conventional technology is shown in FIG. 5. FIG. 6 is a timing chart showing the timing relationship between trigger signals, sampling pulses (charged particle beam pulses) and pass/fail signals in the charged particle beam test system of FIG. 5. FIGS. 4A and 4B are schematic diagram showing sampled waveforms relative to the timings of the sampling pulses.
In the example of FIG. 5, the charged particle beam test system is formed with a charged particle beam tester EBT such as an electron beam tester and an IC tester 40. The charged particle beam tester EBT includes a sampling pulse generator 50, a charged particle beam column 60, a signal processor 70, a test result memory 80, and a controller 90. The charged particle beam column 60 includes a charged particle beam generator such as an electron gun 62, a beam blanker 64, and a secondary electron detector 68. At the bottom of the column 60, an X-Y stage 67 is provided to place a semiconductor device under test (DUT) thereon.
A charged particle beam such as an electron beam emitted from the electron gun 62 is irradiated on the surface of the DUT through the beam blanker 64. The beam blanker 64 controls blanking of the charged particle beam from the electron gun 62 thereby producing a pulsed charged particle beam 61. By moving the X-Y stage 67 in the X and Y directions, the irradiating position of the charged particle beam on the surface of the DUT can be controlled. The signal processor 70 provides a control signal which is provided to the grid 66 to control the amount of secondary electron emitted from the DUT. The secondary electron is detected and integrated by the secondary electron detector 68.
The IC tester 40 provides a test pattern signal 48 with predetermined timings to one or more terminals of the DUT and evaluates the resultant signals from the DUT by comparing the same with the expected value signals. The IC tester 40 also provides a trigger signal 41 to the sampling pulse generator 50 to synchronize the test pattern signal 48 to the DUT with the pulsed charged particle beam 61 irradiated on the DUT. To define the timings of such comparison operations between the DUT's output signal and the expected value signal, the test program used in the IC tester includes several comparator enable (CPE) signals for each time interval between the two trigger signals 41.
The IC tester 40 also sends a fail signal 42 and a clock signal 49 to the charged particle beam tester EBT. The clock signal 49, in this case, is a tester rate clock which defines a test cycle of the IC tester 40. Although not shown in the drawings, a plurality of test cycles are included within each cycle of the trigger signals 41.
In the arrangement of FIG. 5, the IC tester 40 sends the trigger signal 41 to the charged particle beam tester EBT at a predetermined timing relationship with the test pattern signal 48 as shown in FIG. 6A. As noted above, the test program is designed to generate one or more comparator enable signals (CPE) within the trigger cycle as shown in FIG. 6C. Thus, at the timing of the CPE, i.e., for the test cycle having the CPE, the DUT output signal is compared with the expected data signal and a fail signal 42 is generated in the case where the DUT output signal does not match the expected data signal. In the actual test system, the fail signal 42 is produced several cycles later than the test cycle in which the DUT output fails.
The sampling pulse generator 50 includes a delay circuit and generates a sampling pulse 51 at a specified delay time Td after the trigger signal 41 from the IC tester 40. The sampling pulse 51 drives the charged particle beam blanker 64, thus, the column 60 generates the charged particle beam pulse 61 as shown in FIG. 6B. Consequently, the charged particle beam (such as electron beam) pulse 61 is generated at the timing of the sampling pulse 51, which is irradiated on the specified position of the DUT.
In an actual test system, to improve the measurement accuracy, the sampling pulse 51 (beam pulse 61) is repeatedly irradiated on the DUT at the fixed timing Td from the trigger signal 41. For example, the same sampling pulse 51 is repeated by several ten to several thousand times and the data detected by the secondary electron detector 68 is integrated during this period. Namely, the secondary electron detector 68, which is for example, a scintillator or a photomultiplier, converts the detected secondary electron to an electric signal representing the amount of secondary electron and integrates the detected signals for the number of beam pulses supplied to the DUT.
The controller 90 controls the overall process of the charged particle beam tester EBT. For example, the controller 90 sets the delay time Td in the sampling pulse generator 50. After acquiring the number of data for the sampling pulses 51 which is delayed by Td from the trigger signal 41, the controller 90 slightly changes the delay time Td relative to the trigger signal 41 as shown in FIG. 4A to collect the data by repeatedly applying the beam pulse 61 at this timing. By repeating this process while slightly changing the delay time Td, the waveform data such as shown in FIGS. 4A and 4B are acquired and stored in the test result memory 80.
The signal processor receives the trigger signal 41 and the fail signal 42 from the IC tester 40. The signal processor 70 determines that the DUT has failed in the sampling point in question if it receives at least one fail signal 42 before the next trigger signal is received. Based on this pass or fail judgement made by the signal processor 70, the measured data from the secondary electron detector 68 is separately processed and accumulated. The processed measured data 71 separately includes pass potential information (PASS) and fail potential information (FAIL) which is supplied to the test result memory 80.
The signal processor 70 also sends address data to the test result memory 80. The address data is produced based on the trigger signal 41 and the timing relationship between the trigger signal 41 and the delay time Td. In receiving the measured data 71 and the address data, the test result memory 80 stores the pass potential information (PASS) in the PASS memory 82 and the fail potential information (FAIL) in the FAIL memory 84, respectively. Consequently, the potential data of the specified portion of the DUT will be illustrated as a waveform 201 of FIG. 4B.
As noted above, the controller 90 controls the overall process of the charged particle beam tester EBT including the timing shift of the delay time Td. The controller 90 sets various other parameters in the charged particle beam tester EBT as well as communicates with the IC tester to effectively interact with the charged particle beam tester.
In the conventional test system in the foregoing, the plurality of CPE signals are provided in the test program of the IC tester in each cycle of the trigger signal 41 as shown in FIG. 6. If there is a mismatch between the resultant output signal of the DUT and the expected data, a fail signal will be generated by the IC tester. As noted above, the signal processor 70 determines that the DUT has failed if it receives at least one fail signal 42. Thus, if a fail signal 42 is generated at the timing which is different from the charged beam pulse 61, such as a fail signal 42c in the fourth cycle of FIG. 6D, the signal processor 70 determines that the DUT has failed. Thus, the measured data by the secondary electron detector 68 is recorded in the test result memory as the fail potential information. This fail information is incorrect because the DUT response evaluated by the IC tester at the time when the charged particle beam pulse 61 is actually supplied to the DUT shown in FIG. 6B matches the expected data.
Therefore, in the conventional charged particle beam test system, if a fail signal is generated by the IC tester at the timing other than the irradiation of the charged particle beam pulse 61 upon the DUT, there arises an inconsistency between the measured data by the charged particle beam tester EBT and the pass/fail information by the IC tester. In the above example where the fail signal 42c is generated in the fourth cycle of the trigger signal, the measured data must be treated as pass potential information instead of fail potential information. In such a situation, it is not possible to observe a correct potential image such as a waveform 201 but may obtain an incorrect image such as a waveform 202 of FIG. 4B.
If the device test program is modified to generate only one comparator enable (CPE) at the timing of the sampling pulse, i.e., charged particle beam pulse 61, such an inconsistency of measured data may be avoidable. However, this solution has a disadvantage in that many device programs must be prepared for acquiring the data in the different sampling timings such as when the delay time Td relative to the trigger signal 41 is changed.